{"id":374872,"date":"2026-07-31T20:42:47","date_gmt":"2026-07-31T15:12:47","guid":{"rendered":"https:\/\/ebiztoday.news\/?p=374872"},"modified":"2026-07-31T20:42:47","modified_gmt":"2026-07-31T15:12:47","slug":"physical-ai-reshapes-enterprise-infrastructure","status":"publish","type":"post","link":"https:\/\/ebiztoday.news\/index.php\/2026\/07\/31\/physical-ai-reshapes-enterprise-infrastructure\/","title":{"rendered":"Physical AI reshapes enterprise infrastructure"},"content":{"rendered":"
\n

Physical AI is forcing the technology industry to rethink your entire computing stack.<\/p>\n

Robots, autonomous systems and intelligent devices need economical inference, secure data access and infrastructure that works beyond conventional clouds. Rafay Systems is addressing those demands through orchestration that lets providers offer open models without dedicating entire GPU systems to individual customers, potentially lowering enterprise AI costs<\/a> while protecting sensitive information, in keeping with Haseeb Budhani<\/a> (pictured, right), co-founder and chief executive officer of Rafay Systems Inc.<\/p>\n

\u201cThe serverless part on the infrastructure level, we\u2019ve solved that \u2026 we\u2019ve done for some time,\u201d Budhani said. \u201cNow we’ve got a slice of essentially a confidential GPU. I can now have higher security on top of the GPUs that I have already got, which implies as a provider, I get to monetize every second of my infrastructure. But, for the enterprise, it\u2019s actually a greater deal. They like that because they give the impression of being at the whole cost of ownership, they give the impression of being on the capex they usually don\u2019t have the large money. It\u2019s too expensive, otherwise. This enables them to have a secure solution, which actually delivers a greater price point. That\u2019s what everybody wants.\u201d<\/p>\n

Budhani, Eiman Ebrahimi<\/a> (left), CEO of Protopia AI Inc., and other industry leaders spoke with theCUBE\u2019s John Furrier<\/a> and guest host Howie Xu<\/a>, chief AI & innovation officer of Gen Digital Inc., as a part of theCUBE + NYSE Wired: Robotics & AI Infra Leaders event<\/a>, during an exclusive broadcast on theCUBE, SiliconANGLE Media\u2019s livestreaming studio. They examined the infrastructure, economics and technology required to move AI into the physical world<\/a>. The event builds on theCUBE\u2019s Machina AI Summit<\/a> coverage of robotics moving from demonstrations into production. (* Disclosure below.) <\/em><\/p>\n

Physical AI changes the infrastructure equation<\/h3>\n

The economics behind physical AI begin with tokens. Agentic applications eat much more tokens than easy chat interactions because every latest motion may require previous context to be processed again. That makes token volume, model selection and workload design increasingly essential measures<\/a> of infrastructure demand, explained Max Kan<\/a>, tokenomics technical lead at SemiAnalysis LLC.<\/p>\n

\u201cOn the whole, agentic workloads eat exponentially more tokens than the chat workloads everyone was using a yr ago since it\u2019s all inherently multi-turn,\u201d he said. \u201cAnd each time you ask your agent a follow-up or it uses some tool like web search or perhaps it searches your code base, it has to reprocess all previous tokens in your conversation, which really causes total tokens produced\/processed to go up exponentially.\u201d<\/p>\n

The inference market is splitting into specialized workloads, with prefill demanding compute, while decode is determined by memory capability and bandwidth. Positron AI Inc. is targeting those requirements<\/a> through systems optimized for tokens per dollar and tokens per watt, including air-cooled designs suited to enterprise data centers that can’t support liquid-cooled racks, in keeping with Darren Chien<\/a>, managing director of APAC at Positron AI.<\/p>\n

\u201cWe consider that inference is fundamentally an economics problem,\u201d he said. \u201cWe’d like to ensure that these tokens are generated as cheaply and as widely available to users as possible. The best way we compete is on optimizing the 2 metrics of tokens per dollar and tokens per watt.\u201d<\/p>\n

As physical AI moves into factories, vehicles and distributed environments, infrastructure providers need greater than faster accelerators. Axiado Corp. addresses the resulting demands for power, security and system control through silicon-based platform management that adjusts cooling, frequency and voltage by workload while reducing operational pressure<\/a> on CPUs and GPUs at scale, noted Gopi Sirineni<\/a>, founder, president and CEO of Axiado.<\/p>\n

\u201cWe\u2019re bringing that intelligence right into a silicon level, and it would be offloaded off of your GPUs, CPUs,\u201d he said. \u201cWe\u2019re monitoring the system itself, including the fans, liquid cooling and all that. Primary, we manage the efficiency by managing the fans and fan controls and liquid cooling \u2026 after which, two, we do dynamic frequency and voltage scaling for the platform.\u201d<\/p>\n

Intelligence moves closer to the physical world<\/h3>\n

Physical AI increasingly is determined by compact models that may run locally despite limits on power, memory and connectivity. Liquid AI Inc. is developing customizable foundation models<\/a> for laptops, vehicles and other devices, allowing specialized intelligence to cut back inference costs, improve privacy and deliver faster responses without relying entirely on centralized data centers, identified Ramin Hasani<\/a>, co-founder and CEO of Liquid AI Inc.<\/p>\n

\u201cOne small model will not be going to be generally intelligent,\u201d he told theCUBE. \u201cWhat they’ll do, they might be fine-tuned. You may customize them. The good thing about this thing is that customizing a small model, the price of it will be shockingly low.\u201d<\/p>\n

As intelligence moves onto devices, networking becomes a part of the physical AI application architecture. Aria Networks Inc. applies specialized AI to high-resolution telemetry, helping operators detect failures and performance issues before they disrupt costly workloads. The goal is to strengthen human decision-making<\/a> with timely context, not remove people from network operations, in keeping with Mansour Karam<\/a>, founder and CEO of Aria Networks.<\/p>\n

\u201cIt starts with collecting telemetry. There are two points to collecting telemetry,\u201d he said. \u201cYou should be collecting telemetry on the microsecond resolution. That\u2019s primary. You wish it at the proper resolution. Number two, the network, and we will speak about all different networks in an AI factory, nevertheless it spans very different domains.\u201d<\/p>\n

Enterprise AI is moving beyond chatbots toward computer-use agents that may operate existing software, capture workflows and execute routine tasks without lengthy integrations. H Company is targeting that shift by placing a human-centered intelligence layer<\/a> over legacy systems, helping organizations automate repetitive work while keeping consequential actions under human control, emphasized Gautier Cloix<\/a>, CEO of H Company.<\/p>\n

\u201cWhat\u2019s hard is to do it with the security part, without the danger,\u201d he said. \u201cThat\u2019s the identical thing for enterprise. The best way we built our agents is that they can’t execute irreversible tasks \u2013 sending an email, deleting something, ordering something \u2013 and not using a human validation at first \u2026 that\u2019s where we focus lots of energy, ensuring that deterministically our agents cannot do the flawed things.\u201d<\/p>\n

Latest markets emerge around compute capability<\/h3>\n

AI\u2019s rapid growth is creating demand for financial tools that may manage volatile GPU pricing and availability. Silicon Data and The Compute Exchange Inc. are developing compute benchmarks<\/a> and instruments that would help infrastructure operators hedge falling prices while protecting major AI consumers against rising costs as demand shifts over time, explained Carmen Li<\/a>, founder and CEO of Silicon Data and CEO of The Compute Exchange.<\/p>\n

\u201cI\u2019m helping probably a dozen market participants arrange their compute desks. Persons are actually helping their clients to hedge,\u201d she said. \u201cTake into consideration in case your client, in case you\u2019re a bank, your client might be the hyperscaler or neoclouds, they’ve a protracted exposure, they struggle to assist them manage the futures and volatility for his or her revenues. Or \u2026 your client might be the huge AI company, they\u2019re going to eat lots of GPU tokens. They assist them manage the short exposure.\u201d<\/p>\n

Compute markets have gotten more distributed as organizations mix owned infrastructure with capability from neoclouds and regional providers. San Francisco Compute Co. is constructing a marketplace<\/a> for GPU capability that supports this shift, using secure multi-tenancy to match infrastructure with workload, location and availability needs without exposing customer data, emphasized Alan Butler<\/a>, chief business officer of San Francisco Compute Co.<\/p>\n

\u201cThat was the genesis of getting a compute layer where you may rent out on short-term provisioning that became then the premise of, in essence, a marketplace,\u201d he said. \u201cThere\u2019s lots of marketplace corporations which are flipping GPUs over a fence with no SLAs behind it. And the whole lot we do has an SLA standing behind it with an information center provider.\u201d<\/p>\n

Edge deployments require infrastructure that runs models near where data is created without forcing enterprises to rebuild data centers. Axelera AI B.V. is extending its edge-efficient architecture<\/a> into servers and cloud systems, using familiar software frameworks so developers can adopt latest hardware while avoiding proprietary application redesigns and reducing deployment barriers.<\/p>\n

\u201cThey\u2019re designed together very intentionally,\u201d said Alexis Crowell<\/a>, chief marketing officer and general manager of the Americas at\u00a0Axelera AI. \u201cWe have now as many of us working on the software side of our business as we do on the hardware side. It\u2019s not that I\u2019m attempting to monetize the software. I\u2019m attempting to \u2026 ensure that developers have a frictionless experience.\u201d<\/p>\n

To observe more of theCUBE\u2019s coverage of theCUBE + NYSE Wired: Robotics & AI Infra Leaders event<\/a>, here\u2019s our complete video playlist<\/a>:<\/p>\n

https:\/\/www.youtube.com\/watch?v=videoseries<\/a><\/p>\n

(* Disclosure: Neither ScaleFlux Inc., the presenting sponsor of theCUBE + NYSE Wired: Robotics & AI Infra Leaders event, nor other sponsors have editorial control over content on theCUBE or SiliconANGLE.)<\/em><\/p>\n

Photo: SiliconANGLE<\/h5>\n
\n
\n

Support our mission to maintain content open and free by engaging with theCUBE community. Join theCUBE\u2019s Alumni Trust Network<\/strong>, where technology leaders connect, share intelligence and create opportunities.<\/p>\n